EVAPORATION 387 seriously with the operation of a hot-air dryer, and for this reason dryers have been designed that make use of the same air over and over again by letting the excess vapor escape into the atmosphere or condense the moisture by cooling the mixture of air and vapors. The air is usually heated by exhaust or live steam, and as explained in the previous pages on evaporation, the heat transmission depends on: (1) The temperature difference between air and steam; (2) the velocity of the air; (3) the amount of heating surface; (4) the condition of the heating surface, i.e., the material and the diameter of the pipes or heater sections. Small pipes have a higher capacity than larger ones. According to Marlow, page 84, the greatest economical velocity of the air through the heater and dryer is about 20 ft. per second, and with air entering at 60°F., and steam at 220°, the heat transmission, K0, will vary from 1.4 to 1.8 B.t.u. per square foot per hour for 1° temperature difference. Under these conditions of 160° temperature difference, cast-iron radiators will absorb about 260 B.t.u.; 2 to 4 in..wrought-iron pipes, 300 B.t.u.; and 1 to 2 in. pipes, 350 B.t.u. Carrier (Trans. A. S. M. E., Vol. 33) gives for higher steam pressure (50 lb.)K0 = l/(a -f (b/w)), in which a and b are constants (a = 0.045 and b = 55 average), and w is the velocity of air in feet per minute. This equation gives much higher values for K0 than stated by Marlow. Practical data on this subject are given in the catalogs of heater manufacturers. The exact determination of the amount of air and gases needed for the drying of a certain substance requires rather intricate calculations. In practically all cases, one must consider constants and factors that are determined by experience and actual tests only. However, in all dryer installations of this kind, the following points are important: (1) The quantity of hot air or gases required to raise the temperature of the substance, make up all losses, and vaporize the moisture; (2) the amount of air required to carry off the vapors after the temperature of the air has been lowered by the heating of the material and evaporation of the moisture; (3) the air and gases must never be heated to such a temperature as will injure or ignite the material; (4) the discharge temperature of the air must be such that the vapors will not condense by coming in contact with the material; (5) high temperatures of the incoming and outgoing air will increase the capacity as they will raise the temperature difference between heating medium and material, and also the carrying power of the air, but they will decrease the efficiency; (6) the fuel or steam consumption of a dryer is lowest when the outgoing air or gas is discharged at a temperature of about 176°F., and a saturation of from 80 to 90 per cent, and under otherwise equal conditions the efficiency is practically the same: (a) In a dryer where hot gases enter at 1,200° and more, and are discharged at 170° and 80 per cent saturation; (b) in a dryer where the air is used over and over again and only a part is discharged at 170° and 80 per cent saturation; (c) in a dryer which consists of several individual chambers, and where the air is re-heated repeatedly but finally discharged at 170° and 80 per cent saturation: (7) wherever possible, the heating surface should be placed inside the dryer to avoid losses and the waste air and gases must be removed as quickly as possible. The location of the discharge opening is also important and should be determined by calculation or experience. More complete information, tables and calculations of capacities and efficiencies of such dryers will be found in the books of Hausbrand and Marr. All the data and constants should be used with caution as even a slight change in the process of manufacture will influence the conditions under which the material may be treated in a dryer. For this reason the manufacturer of drying machinery will usually ask for a sample of the material to be dried, and will only submit proposals after laboratory experiments and tests.